BMC Microbiology

Table 2 Performance of trained ML models with random seeds

From: Machine learning and phylogenetic analysis allow for predicting antibiotic resistance in M. tuberculosis

	Support Vector Machine (SVM)				Random Forest
	Hold-out		Cross Validation		Hold-out		Cross Validation
	MEAN	STDEV	MEAN	STDEV	MEAN	STDEV	MEAN	STDEV
MCC
AMK	0.775	0.034	0.771	0.057	0.859	0.026	0.853	0.059
CAP	0.671	0.061	0.661	0.060	0.766	0.035	0.799	0.040
ETH	0.367	0.072	0.353	0.068	0.519	0.054	0.566	0.074
KAN	0.688	0.031	0.686	0.047	0.820	0.044	0.866	0.025
OFL	0.555	0.037	0.571	0.068	0.801	0.035	0.809	0.024
STR	0.626	0.026	0.611	0.030	0.799	0.013	0.816	0.018
ROCAUC
AMK	0.873	0.021	0.874	0.031	0.904	0.027	0.895	0.042
CAP	0.843	0.028	0.849	0.034	0.865	0.012	0.885	0.023
ETH	0.685	0.039	0.677	0.036	0.760	0.027	0.777	0.031
KAN	0.850	0.021	0.853	0.033	0.889	0.023	0.919	0.019
OFL	0.780	0.018	0.786	0.032	0.887	0.015	0.891	0.014
STR	0.821	0.013	0.815	0.013	0.892	0.007	0.898	0.013

Back to article page

ISSN: 1471-2180

Contact us

General enquiries: journalsubmissions@springernature.com